iCASE In silico characterisation of portal proteins for application as biosensors

由牛津纳米孔技术公司(ONT)首创的纳米孔测序是基于DNA穿透工程版本的CsgG，CsgG是一种来自大肠杆菌的分泌蛋白质。电场诱导DNA穿过嵌入在不透膜中的孔状门户蛋白，产生由仪器记录的离子电流的减少。电流图谱是通过孔的核苷酸碱基的特征，使得能够识别bas序列。与其他测序方法相比，纳米孔碱基测序已被证明具有多方面的优势，包括便携性、实时数据采集、长和超长读数分析、直接RNA测序和碱基修饰的检测。纳米孔技术的范围具有巨大的潜力，可以扩展到其他生物医学应用，包括蛋白质测序、翻译后修饰的检测、毒素、病毒颗粒和其他分析物的原位检测。该项目旨在推动优势纳米孔生物传感的进一步发展，为分子和细胞医学开辟新的前景。多功能生物传感需要设计出具有可行特性的新门户蛋白，如有效的膜嵌入、在检测条件下的稳定性和适当的孔-分析物互补。这项研究将应用计算方法来指导合理优化病毒门户蛋白的物理和化学性质，从而使其应用于生物传感器。将结合几种计算技术来对候选蛋白质进行表征。经典分子动力学(原子论和粗粒度)将与量子力学计算一起使用，以提供彻底的、预测性的表征。此外，利用AlphaFold2等与蛋白质相关的深度学习模型的最新进展，将促进蛋白质序列的优化。计算实验将由ONT完成的对拟议的蛋白质修饰的实验评估反复支持。该项目将使学生能够发展量化专业知识，以及广泛的技能，以解决跨学科研究和创新。

Nanopore sequencing pioneered by Oxford Nanopore Technologies (ONT) is based on DNA threading through an engineered version of CsgG, a secretion protein from E. coli . Electric-field induced threading of DNA through the pore-shaped portal protein imbedded within an impermeable membrane produces the reduction in ionic current recorded by the apparatus. The current profile is characteristic to the nucleotide bases passing through the pore, enabling identification of the bas sequence. Nanopore bases sequencing has already proved to have multiple strengths including portability, real-time data acquisition, analysis of long and ultra-long reads, direct RNA sequencing and detection of base-modifications over other sequencing methodologies.The scope of nanopore technology has great potential to be extended towards other biomedical applications including protein sequencing, detection of posttranslational modifications, in situ detection of toxins, viral particles and other analytes. This project aims to drive further development of advantageous nanopore biosensing opening new perspectives in molecular and cellular medicine.Versatile biosensing requires engineering of new portal proteins with feasible properties such as effective membrane embedding, stability under the assay conditions, and proper pore-analyte complementarity. This research will apply computational methods to guide rational optimisation of physical and chemical properties of viral portal proteins allowing their application as biosensors. Several computational techniques will be combined to perform characterization of candidate proteins. Classical molecular dynamics (atomistic and coarse-grained) will be used alongside quantum mechanics calculations to provide thorough, predictive characterization. In addition, protein sequence optimisation will be facilitated by harnessing the recent progress in protein-related Deep Learning models such as AlphaFold2 . The computational experiments will be iteratively backed up by experimental assessment of proposed protein modifications completed by ONT.The project will enable the student to develop quantitative expertise, as well as a broad range of skills to address interdisciplinary research and innovation.